Multi-scale ecological filters shape the crayfish … 1 Major article: Multi-scale ecological...

1

Major article: Multi-scale ecological filters shape the crayfish microbiome 1

Authors: James Skelton*1,6, Kevin M. Geyer2,7, Jay T. Lennon3,8, Robert P. Creed4,9, and Bryan 2

L. Brown5,10 3

1 School of Forest Resources and Conservation, University of Florida, Gainesville, Florida 32603 4

USA 5

2 College of Life Sciences and Agriculture, University of New Hampshire, Durham, New 6

Hampshire 03824 USA 7

3 Department of Biology, Indiana University, Bloomington, Indiana 47405 USA 8

4 Department of Biology, Appalachian State University, 575 Rivers Street, Boone, North 9

Carolina 28608 USA 10

5 Department of Biological Sciences, Virginia Polytechnic Institute and State University, 11

Blacksburg, Virginia 24061 USA. 12

Email: 6 [email protected], 7 [email protected], 8 [email protected], 9 13

[email protected], 10 [email protected] 14

Running headline: The crayfish microbiome 15

16

17

PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.2266v1 | CC BY 4.0 Open Access | rec: 8 Jul 2016, publ: 8 Jul 2016

2

Abstract 18

1) Communities of symbiotic microbes obtained from the environment are an integral component 19

of animal fitness and ecology. Thus a general and practical understanding of the processes that 20

drive microbiome assembly and structure are paramount to understanding animal ecology, 21

health, and evolution. 22

2) We conceptualized a series of ecological filters that operate at the environment, host, and host 23

tissue levels during microbiome assembly and discuss key ecological processes that structure 24

animal microbiomes at each level. 25

3) We conducted a survey of crayfish across four sites within the contiguous range of the of 26

stream-inhabiting crayfish Cambarus sciotensis in western Virginia, USA, to characterize 27

multiscale variation in the crayfish microbiome. We also conducted an in situ experiment to 28

assess local drivers of microbial diversity on the closely related Cambarus chasmodactylus. We 29

used a combination of DNA fingerprinting and next-generation sequencing to characterize 30

microbiome diversity and composition from crayfish carapaces and gills to identify key filters 31

affecting microbiome structure. 32

4) Field survey showed that local environment and host tissues interact to create patterns of 33

microbial diversity and composition, but the strongest effects on microbial community structure 34

were observed at the level of host tissue. Our field experiment confirmed strong effects of host 35

tissue, and also showed that a metazoan ectosymbiont which feeds on biofilms (Annelida; 36

Branchiobdellida) had significant effects on microbial composition of the host carapace. Crayfish 37

carapaces were colonized by diverse and taxonomically even microbial communities that were 38

similar to, and correlated with, microbial communities of the ambient environment. Conversely, 39


3

crayfish gills were colonized by less diverse communities and dominated by two families of 40

bacteria with potentially significant functional roles: Comamonadaceae and Chitinophagaceae. 41

5) Our results suggest that microbial assembly of the carapace is driven by external biotic and 42

abiotic processes, whereas assembly on the gills appears to be coupled to host biology that favors 43

interactions with few specific taxa. Our work shows how multi-scale studies of symbiont 44

community assembly provide valuable insights into how the animal microbiome is structured 45

under conditions of natural complexity and help identify other symbiont taxa, i.e., the 46

branchiobdellidans, that may further influence microbiome assembly and structure. 47

Key-words: bacteria, branchiobdellida, cleaning symbiosis, defensive symbiosis, disease 48

ecology, metacommunities, parasites 49

50


4

Introduction 51

Recent advances in our understanding of the intimate and diverse interactions between 52

animals and their associated microbiota have blurred the lines that define individual organisms 53

and fundamentally changed the way we think about organismal biology, ecology, and evolution 54

(Gilbert, Sapp & Tauber 2012). Simultaneously, observational studies utilizing new sequencing 55

technologies have highlighted patterns of symbiont microbial diversity and composition across a 56

range of contexts from specific host body parts to geographic regions (e.g. Costello et al. 2009; 57

Dominguez-Bello et al. 2010; Kuczynski et al. 2010). Increasingly, frameworks that embrace the 58

multi-scale processes of community structure are being used to explain symbiont diversity (e.g. 59

Dethlefsen, McFall-Ngai & Relman 2007; Pedersen & Fenton 2007; Graham 2008; Mihaljevic 60

2012). Though promising, these frameworks are still in need of empirical evaluation, and 61

particularly experimental evaluation. 62

Mounting evidence is showing that functionally significant relationships with microbial 63

symbionts are ubiquitous among metazoans (Scheuring & Yu 2012; Clay 2014). While the 64

attention payed to animal microbiome research has been biased towards terrestrial vertebrates, 65

especially humans, microbial symbionts are important in other host-associated systems. For 66

instance, skin microbiota are thought to serve as protective agents against amphibian pathogens 67

(Harris et al. 2009; Loudon et al. 2014). By unlocking novel food resources, or competitively 68

excluding potentially harmful pathogens beneficial microbial symbionts are essential to a diverse 69

and growing list of aquatic invertebrates as well, including sponges, corals, and arthropods 70

(Scheuring & Yu 2012; Peerakietkhajorn et al. 2015). Thus important and yet undiscovered 71

relationships between crayfish and microbial symbionts seem likely. Crayfish have a world-wide 72

distribution in freshwaters, serve as keystone species and ecosystem engineers (e.g. Creed 1994; 73


5

Statzner et al. 2000; Usio & Townsend 2002; Statzner, Peltret & Tomanova 2003; Creed & Reed 74

2004; Usio & Townsend 2004) , comprise many threatened and endangered species (Taylor et al. 75

2007; Helms et al. 2013; Owen et al. 2015), as well as several globally invasive nuisance 76

species, and are one of the most frequently used model organisms in organismal biology 77

(Holdich & Crandall 2002). Part of the interest in crayfish research stems from their amenability 78

to field and laboratory studies, including experimental studies of symbiotic associations (Skelton 79

et al. 2013). Despite the wealth of interest and intense study of crayfish, we are unaware of 80

previous investigations focused on the crayfish microbiome. 81

In this study, we conceptualized the bacterial microbiome of stream inhabiting crayfish as 82

the result of a series of nested filters operating at scales from the geographic region, to the host 83

tissues (microsite community, Figure 1). At the coarsest scale, physical and biological 84

characteristics of a local environment filter out a subset of all bacterial taxa that could potentially 85

colonize the habitat from the regional species pool (Figure 1 – a). This subset of the regional 86

pool represents the environmental pool of potential microbial symbionts for hosts living in a 87

given local habitat. Similarly, it is likely that only a subset of the environmental pool is suited for 88

life in symbiosis with a particular host and thus microbiome community membership is filtered 89

at the interface between host and environment (Figure 1 – b). Furthermore, variation in microbial 90

habitat quality and interactions among symbionts that occur in or on the tissues that form 91

different parts of the host body, here called “microsites”, may further filter colonizing microbial 92

symbionts (Figure 1 – c). Under this framework of nested filters, we implemented field surveys 93

and an in-situ experiment to characterize the bacterial microbiome of crayfish and explore multi-94

level drivers of diversity. We examined the strength of environmental filtering on microbial 95

symbiont communities by comparing environmental and symbiotic bacterial communities from 96


6

four watersheds. We then relate environmental patterns to the patterns observed in the 97

microbiome of crayfish from each site. To asses environmental filtering at the level of host 98

tissue, we compared microbial community patterns observed on tissues which are highly 99

interactive with crayfish physiology and one that is less interactive; the gills and carapace, 100

respectively. Finally, we conducted a field experiment with manipulated co-infection treatments 101

of Cambarincola ingen (Annelida: Branchiobdellida; Hoffman, 1963), an obligate ectosymbiont 102

that feeds on crayfish biofilms, to explore the effects of symbiont-symbiont interactions on the 103

diversity and composition of the crayfish microbiome. Our results demonstrate the necessity of a 104

complete multi-level framework to understand patterns of microbial symbiont diversity and to 105

identify key interactions that may influence microbiome assembly and structure. 106

Materials and methods 107

Field survey: We examined the crayfish microbiome from 4 sampling sites within the New River 108

drainage near Blacksburg, Virginia, USA; Sinking Creek (37°18’9.34’’ N, 80°29’6.9’’ W), Big 109

Stoney Creek (37°24’53.33’’ N, 80°34’53.58’’ W), Tom’s Creek (37°14’23.82’’ N, 110

80°27’30.55’’ W), and Spruce Run Creek (37°15’54.43’’ N, 80°35’52.67’’ W). Six adult 111

crayfish (Cambarus sciotensis; Rhoades, 1944) of similar size (35 mm mean carapace length 112

[CL], ± 5 mm SD) from each site were chosen from which to collect gill and carapace biofilm 113

samples. Terminal restriction fragment length polymorphism (TRFLP) analysis of the 16S rRNA 114

gene was used to estimate bacterial diversity and compositional variation (Thies 2007) of these 115

samples, in addition to three replicate samples of benthic substrate biofilms from each site. 116

TRFLP is a largely automated process suited for high sample through-put and is useful for 117

tracking changes in microbial community structure at coarse taxonomic scales over time and 118

space (Schutte et al. 2008). We limited our sampling to the crayfish species C. sciotensis of 119


7

similar size to minimize effects of host age on microbiome composition; 35 mm mean carapace 120

length (CL), ± 5 mm SD. All sampling sites were within the contiguous range of this species 121

(Hobbs, Holt & Walton 1967). Sampling of benthic substrate consisted of swabbing a 1 cm2 122

upward facing surface of randomly selected cobbles from the streambed. 123

Cleaner symbiont experiment: We conducted a field experiment to specifically characterize the 124

microbial communities of the crayfish gills and carapace, and to assess the influence of metazoan 125

ectosymbionts on the crayfish microbiome. Twenty crayfish enclosures (“cages”) were installed 126

in the South Fork of the New River near the campus of Appalachian State University in Boone, 127

N.C. in June 2012. Cages were constructed of welded aluminum frames with solid aluminum 128

sides and bottoms. Two layers of hardware clothe mesh (12 mm) were on the front and backs of 129

the cages permitted unimpeded flow of stream water while isolating individual crayfish for 130

recapture and preventing transmission of ectosymbionts to caged crayfish from external crayfish 131

(Brown et al. 2012). Steel bars (rebar, 2 per side) were driven into the streambed on both sides 132

of each cage to hold it in place during against high discharge events. Approximately 40 L of 133

mixed substrate collected immediately downstream was added through hinged, mesh-covered 134

lids on top. Cages were arrayed according to a randomized block design, with 5 blocks 135

perpendicular to stream flow to allow estimation of upstream/downstream effects of cage 136

placement. Each block contained four cages, wherein a single crayfish received one of four 137

treatments: no worms (all branchiobdellidans removed), 6 worms, 12 worms, or immediate 138

harvesting for assessment of initial microbial communities at the experiment’s outset. 139

Crayfish (Cambarus chasmodactylus; James, 1966) were collected from nearby 140

tributaries of the New River, and chosen based on carapace length (30-35mm), and presence of 141

all their appendages. All crayfish were cleaned of branchiobdellidan worms in the laboratory via 142


8

manual removal followed by 5 min submersion in a 0.5M MgCl2 solution (Brown et al. 2002). 143

Crayfish were then immediately relocated to experimental cages to allow reconditioning of gill 144

and carapace surfaces with native microorganisms. After 13 days, we began the experiment by 145

transplanting large worms (C. ingens) collected from local tributaries onto crayfish at treatment 146

levels. Midway through the 43 d experiment, crayfish were physically examined and worm 147

treatments adjusted by reapplying missing worms as necessary to achieve initial treatment levels. 148

Debris was removed from the exterior of cages every other day throughout this period to prevent 149

accumulation of sediment or altered flow in/around cages. 150

Microbial sampling protocols: Sampling of microorganisms was performed in-field using flame-151

sterilized equipment. Carapace communities were sampled by wiping one randomly selected 152

lateral half of the carapace with a sterile swab for 10s and immediately preserving the swab head 153

in a microcentrifuge tube preloaded with a sucrose lysis buffer (pH = 9) (Mitchell and Takacs-154

Vesbach 2008). Similar techniques were used to sample a 1 cm2 area of cobbles comprising the 155

benthic substrate of our field survey. For gill samples, the carapace was removed and the rear-156

most gill tuft from one lateral side (randomly chosen) was clipped and stored in a tube of buffer. 157

All samples were kept on ice and moved to -80°C storage within hours. All remaining crayfish 158

carcasses were preserved in 70% ethanol for later observation. 159

DNA extraction from gill and carapace samples was performed using a conservative 160

technique optimized for potentially low-biomass and low diversity environmental samples 161

(Geyer et al. 2013). Briefly, DNA was extracted using a cetyltrimethylammonium bromide 162

(CTAB) procedure that involves a mixture of 1% CTAB, 10% sodium dodecyl sulfate, 163

phenol/chloroform/isoamyl alcohol (pH=7.5), lysozyme (0.2µg/µL), and proteinase K (20µg/µL) 164

with either a swab head or gill tissue. Extracted DNA was resuspended in Tris buffer (pH=8.0) 165


9

and quantified via spectrophotometry (NanoDrop 2000; Thermo Scientific, Wilmington, DE, 166

USA). The average recovery of DNA from carapace swabs was ~50ng DNA/μL and ~500-167

1000ng DNA/μL per gill filament (a majority of which was assumed to be crayfish DNA). 168

TRFLP Analysis: PCR amplification of extracted DNA took place in triplicate (25µL reaction 169

volume) using a standard 2uL of diluted template, 5 units/µL of Taq Hot Start Polymerase 170

(Promega Corporation, Madison, WI, USA), and the universal bacterial primers 8F (5'-171

AGAGTTTGATCMTGGCTCAG-3') and 519R (5'-ACCGCGGCTGCTGGCAC-3'), the forward 172

primer labeled with a 5' 6-FAM fluorophore (Integrated DNA Technologies, Coralville, IA, 173

USA). Amplification was optimized for concentrations of MgCl2 (2.5mM per reaction), BSA 174

(1µL/reaction), annealing temperature (53°C), and final extension time (5min). Amplification 175

replicates were pooled and cleaned using a QuickClean II PCR Extraction Kit (GenScript, 176

Piscataway, NJ, USA). Amplifications were digested with HaeIII (New England BioLabs, 177

Ipswich, MA, USA) in triplicate (20µL reaction volume) for 3 hours at 37°C following 178

manufacturer's suggested protocols. Digestion replicates were then pooled and cleaned using 179

GenScript extraction kits. Fragment separation/quantification took place in quadruplicate with 180

an ABI 3130xl Genetic Analyzer (Applied Biosystems, Carlsbad, CA, USA) and fragments 181

binned using the GeneMarker software AFLP protocol. Resulting sample profiles were 182

standardized using the procedures outlined in Dunbar (2001) to produce both a consensus profile 183

among replicates and final normalization of all sample profiles by total sample fluorescence. 184

Next-generation Sequencing: Illumina MiSEQ amplicon sequencing was used to examine the 185

diversity of bacterial communities. Approximately 150ng of 48 DNA extracts were provided to 186

the Virginia Bioinformatics Institute at Virginia Tech for 150bp paired end sequencing of the 187

V4-V6 region of the16S rRNA bacterial gene. The result was ~500k reads per sample with an 188


10

average length of 253bp after stitching of paired reads. PANDAseq was used to merge forward 189

and reverse reads, correct errors in the region of overlap, and reject any reads that failed to 190

overlap sufficiently (Masella et al. 2012). Taxonomy was assigned to all unique sequences using 191

the uclust method and GreenGenes reference database (13_8 release) (DeSantis et al. 2006) in 192

QIIME (1.7.0) (Caporaso et al. 2010). USEARCH was used to cluster all dereplicated reads into 193

OTUs at the 97% similarity level and filter chimeras (Edgar 2010). The UPARSE pipeline was 194

followed for all data denoising. (Edgar 2013). A total of 10,489 OTUs were successfully 195

annotated and formed the basis of multivariate analysis. 196

Statistical methods: To assess bacterial diversity observed on carapaces and gills, we calculated 197

Simpson diversity for each sample from normalized TRFLP fragment peaks using the diversity() 198

function of the vegan package for R v2.0 – 10 (Oksanen et al. 2013). Simpson diversity indices 199

were converted to effective numbers of taxa following Jost (2006). This conversion 200

approximates the total number of equally abundant taxa in a community and makes comparisons 201

among communities more interpretable and intuitive than raw diversity indices (Jost 2006). We 202

used linear regression to correlate the observed diversity of crayfish samples with bacterial 203

diversity of the local substrate and to compare diversity observed on gills to that of carapaces. 204

We first calculated the average diversity of all substrate samples from each sampling locality, 205

and then used average substrate diversity as a continuous predictor and crayfish tissue (gill 206

versus carapace) as a categorical factor, with an interaction term. We also used linear regression 207

to assess the effects of sample type (gills versus carapaces) and branchiobdellidan treatment level 208

with an interaction term in our symbiont experiment. In this case, diversity was calculated as 209

described above using the number of reads obtained from each OUT obtained from 210

bioinformatics of NGS sequence data. 211


11

We examined the main and interactive effects of host tissue (gills versus carapaces) and 212

sampling site on the TRFLP fragment composition (a proxy for bacterial community 213

composition) using permutations multivariate analysis of variance (PERMANOVA; Anderson 214

2001) implemented using the adonis() function in the R package Vegan v2.0 - 10 (Oksanen et al. 215

2013), with a Bray-Curtis dissimilarity matrix and 10,000 permutations. Compositional effects 216

were visualized by non-metric multidimensional scaling (NMDS) using the metaMDS() function 217

in vegan v2.0 – 10. Similar methods were used to assess the effects of branchiobdellidan 218

treatment levels on the bacterial composition of gills and carapaces in the symbiont experiment. 219

We also used Mantel tests to assess the strength of multi-variate correlations between average 220

within-site TRFLP profiles from substrate samples, to those of gills and carapaces across all 4 221

sites. This was done by taking the average peak height for each fragment length from each 222

sample type (gills, carapaces, and cobbles) within each site to create 3 site by fragment length 223

matrices; 1 matrix each for cobbles, gills and carapaces. We then used 2 Mantel correlations to 224

correlate gills and cobble, and carapaces and cobbles across sites (mantel() function; vegan 2.0 – 225

10; Oksanen et al. 2013). 226

227

Results 228

Field survey: TRFP results recovered an average of 37.9 (min = 21, max = 47) and 20.3 (min = 229

8, max = 36) OTU taxa from carapace and gill tissues, respectively, suggesting greater overall 230

phylum-level richness of carapace biofilms (Fig 2). Cobbles sampled from the stream bed had an 231

average of 30.2 taxa (min = 13, max = 50). There was considerable variation in microbial 232

diversity within host tissues, among sampling locations (Fig 2), and marginally significant, 233

interactive effects of environmental (non-host) microbial diversity and host microsite on 234


12

symbiotic microbial diversity (Table 1). Among-site variation in microbial diversity of the 235

carapace was correlated with the microbial diversity of local substrate, however among-site 236

variation in the diversity of the gills was not related to substrate diversity (Fig 2). 237

Host microsite accounted for most of the compositional variation among samples 238

(PERMANOVA, F1,33 = 23.86, p < 0.001, R2 = 0.33). Additionally, there was significant among-239

site variation in microbial composition (PERMANOVA, F3,33 = 2.68, p = 0.003, R2 = 0.11), and 240

interactive effects of sampling site and microsite on composition (PERMANOVA, F3,33 = 2.54, p 241

= 0.003, R2 = 0.10). NMDS ordination recovered 2 convergent solutions after 12 tries and a final 242

stress of 0.12 (Fig 3). The bacterial composition of the carapace biofilm was largely 243

indistinguishable from that of local substrate biofilms, whereas the composition of the gills was 244

distinct from the substrate (Fig 3). Across sampling sites, there was a significant correlation 245

between the bacterial composition of the benthic substrate and the carapace microbiome (Mantel 246

r = 0.812, p = 0.034) and a marginally significant correlation between substrate composition and 247

the composition of the gills biofilm (Mantel r = 0.527, p = 0.089). 248

Cleaner symbiont experiment: Similar to the TRFLP results from the field survey, Illumina 249

sequencing of the experimental crayfish biofilms confirmed that the bacterial communities of 250

carapaces were much more diverse (mean = 35.9) than those of the gills (mean = 4.76). There 251

were no significant effects of branchiobdellidan treatments on the microbial diversity of the 252

carapace or the gills (Table 2). Again, similar to the results of our field survey, the microbial 253

composition of the carapace was distinct from that of the gills. Bacterial communities of the 254

carapaces contained a large number of evenly abundant taxa, whereas the gills contained many 255

fewer taxa and were dominated by two bacterial families; Comamonadacea and Chitinophagacea 256

(Fig 4). Comamonadacea had the highest relative abundance of all taxa present in gill samples, 257


13

comprising an average of 41.0% (± 3.9% SE) of reads in gill samples. Second was 258

Chitinophagacea which comprised an average of 15.5% (± 2.4% SE) of reads from gill samples. 259

Conversely, these taxa comprised and average of 4.0% (±0.007% SE) and 0.4% (± 0.002% SE) 260

of reads from carapace samples respectively. Branchiobdellidan treatments had no detectable 261

effects on the microbial composition of the crayfish gills (PERMANOVA F2,12 = 0.732, p = 262

0.782). In contrast, there was a significant effect of worm treatment on the bacterial composition 263

of the carapaces in which crayfish exposed to high symbiont densities were distinct from controls 264

and medium density treatments (Fig 5; PERMANOVA F2,16 = 1.646, p = 0.007, R2 = 0.17). 265

Discussion 266

Although some microbial symbionts are transmitted vertically from parent to offspring, 267

or horizontally from host to host, most are obtained from environmental sources (Bright & 268

Bulgheresi 2010; Walke et al. 2014). In this study, we examined ecological filters that operate at 269

multiple levels as the crayfish microbiome is assembled from environmental sources. We found 270

significant sources of variation at each level, from geographic sampling location, to interactions 271

among metazoan and microbial symbionts within specific microsites. By far, the strongest effects 272

were observed at the tissue-specific interface between host body and environment. 273

Host habitat level: In our field survey we found correlative evidence for environmental controls 274

over the crayfish microbiome. We observed significant variation in microbial composition 275

among sampling sites, and correlations between the microbial composition of environmental 276

samples and composition of the gill and carapace symbiont communities. Similarly, microbial 277

diversity of environmental samples was positively correlated with carapace communities, but 278

environmental and gill communities were not correlated. Thus, the taxonomic composition of 279

bacteria on the gills and carapace are influenced to some degree by the local environment. 280


14

However, only carapace bacterial diversity was correlated with local environment, and bacterial 281

diversity of the gills showed no relationship with environmental diversity. This result suggests 282

that a process acting at a finer level limits membership of the gills, and gill community diversity, 283

but not carapace diversity, was saturated at all sampling sites. We suspect that the observed 284

variation among sites in both environmental and crayfish samples may be attributed to physical 285

characteristics of the sampling sites that influence environmental microbial species pools. For 286

instance, global patterns in soil and stream bacteria composition and diversity are shaped by pH 287

gradients at both world-wide and local spatial scales (Fierer et al. 2007; Lauber et al. 2009; 288

Rousk et al. 2010). Although we did not assess physiochemical characteristics of our sampling 289

sites in this study, pH seems to be a likely underlying environmental driver of among site 290

variation based on the results of work from another temperate watershed in the eastern United 291

States (Fierer et al. 2007). 292

Host-habitat interface: As microbial symbionts are acquired from the surrounding environment, 293

membership of the microbiome may again be filtered at the interface of host and environment. 294

Not all microorganisms present in an environment are suited for symbioses with an animal host 295

and variation among host species and individuals select for different subsets of the 296

environmental pool (McKenzie et al. 2012; Kueneman et al. 2014). Ecological and host-297

mediated differences among microsites on or in the host body may select for a more specialized 298

subset of the global pool. For example, microbial communities of human skin vary widely 299

among parts of the human body (Costello et al. 2009; Grice & Segre 2011). Therefore the 300

communities of bacteria found in association with a host tissue may represent a heavily restricted 301

subset of the global pool of potential colonists. Our results suggest that filtering is strong at the 302

gill microsite, but weak or nonexistent on the carapace. Carapace diversity was comparable, or 303


15

even greater than bacterial samples taken from local substrates. Conversely, microbial 304

communities of the gills were shown by TRFLP analysis of survey data to be less diverse than 305

local substrate and carapaces at all sampling sites, and again found to be less diverse than 306

carapace biofilms by direct sequencing of experimental animals. Thus the strong tissue-specific 307

filtering that occurs at the gills imposes a strong filter on microbial symbiont communities, but 308

similar filtering is weak or non-existent on the less physiologically active carapce. 309

Microsite processes: We predicted that the microbial communities of more inert tissues would be 310

largely composed of opportunistic, environmental symbionts, whereas the microorganisms 311

colonizing more active tissues would include specialist symbionts. Our results support this 312

prediction. Crayfish gills are a vital interface between the crayfishes’ internal physiology and 313

external environment. The gills exchange respiratory gases with the environment, are sites of 314

nitrogenous waste excretion, and regulate ion exchange. Consequently, the gills are composed of 315

thin, un-sclerotized epithelia. Conversely, the carapace is composed of heavily sclerotized and 316

calcified cuticle, and is approximately 300 times less permeable than the gills (Pequeux 1995). 317

Our field survey suggest that the microbial communities of the crayfish carapace are similar to 318

the surrounding substrate in terms of both diversity and composition. Moreover, variation among 319

sampling sites revealed correlations between bacterial communities of local substrate and the 320

crayfish carapace, indicating that the microbiota of the crayfish are probably composed of mostly 321

opportunistic taxa from the surrounding environment. Congruent to survey of DNA fingerprint 322

data, sequence data from our field experiment revealed relatively even, highly diverse 323

communities composed of ubiquitous environmental taxa. In contrast to the carapace, crayfish 324

gills were found to have less diverse microbial communities that were dominated by few taxa. 325

While carapaces were characterized by many evenly distributed taxa, the gills were dominated 326


16

by two bacterial families: Comamonadacea and Chitinophagacea. Reduced bacterial diversity, 327

and constrained microbial membership of the crayfish gills may be at least in part attributable to 328

antimicrobial agents with crayfish hemolymph. Previous work has shown that the hemolymph of 329

some crayfish species can inhibit the growth of some bacterial taxa in culture and may reduce 330

bacterial accumulations on crayfish gills (Farrell, Creed & Brown 2014). 331

Evidence from other freshwater animals suggests that some taxa found in this study may 332

serve a significant function in crayfish biology. Recent experimental work has demonstrated a 333

positive effect of bacterial symbionts on the population growth rates of a small freshwater 334

crustacean, Daphnia (Peerakietkhajorn et al. 2015). Similar to our study, these beneficial 335

microbial communities were largely composed of Betaproteobacteria belonging to the 336

Comamonadacea; a finding congruent to previous investigations of the bacterial symbionts of 337

Daphnia (Qi et al. 2009; Freese & Schink 2011). Moreover, Comamonadacea and 338

Chitinophagacea are often dominant members of the skin microbiomes of aquatic amphibians 339

(Harris et al. 2009; McKenzie et al. 2012; Kueneman et al. 2014; Walke et al. 2015), yet the 340

processes that led to their ubiquity and dominance in aquatic animal microbiomes, and their 341

potential functional significance are not resolved. 342

The numerical dominance of Comamonadacea and Chitinophagacea found here and in 343

other aquatic microbiome suggest they may serve as defensive symbionts. Increasingly, 344

microbiome studies are focused on non-random microbiome community assembly, with a 345

particular focus on beneficial microbial taxa (e.g. Bäckhed et al. 2005; Dethlefsen et al. 2006; 346

Dethlefsen, McFall-Ngai & Relman 2007; Huttenhower et al. 2012; Scheuring & Yu 2012; 347

Shafquat et al. 2014). Most beneficial microbes provide either nutritional advantages or defense 348

against invading pathogens (Bäckhed et al. 2005; Huttenhower et al. 2012). Defensive microbes 349


17

most often protect their hosts by establishing as competitive dominants through interference 350

composition by producing antibacterial compounds (Scheuring & Yu 2012). Recent work 351

suggests that hosts favor the competitive dominance of defensive microbes through resource 352

provisioning (Scheuring & Yu 2012). In this study, we found very little evidence that the 353

microbial communities of the crayfish carapace assemble non-randomly or contain functionally 354

significant taxa. In stark contrast to carapaces, the microbial communities of the gills comprised 355

a very limited set of microbial taxa. From recent theoretical developments (Scheuring & Yu 356

2012), we speculate that microbial resources released through the thin and permeable cuticle of 357

the gills fuels the competitive dominance of Comamonadacea and Chitinophagacea, and that one 358

or both taxa may produce antimicrobial compounds to maintain dominance and protect the host 359

from invading pathogenic taxa. The gill cuticle itself could be a resource for Chitonphagacea, as 360

several strains within this family are known to degrade chitin (Rosenberg 2014). Similar 361

relationships between animals and protective microbes are common in nature and have been 362

described in many marine invertebrates such as sponges and corals (Scheuring & Yu 2012; Clay 363

2014), and are therefore likely to be common freshwater invertebrates. Though the nature of the 364

symbioses between Comamonadacea, Chitinophagacea, and their hosts are currently speculative, 365

these relationships deserve further attention. 366

The study of symbioses continues to expand beyond concepts based on pairwise species 367

interactions and is embracing the realistic complexity of symbiosis. However, perspectives that 368

transcend microbe-microbe or microbe-host interactions are not typically considered. Recent 369

synthetic work has highlighted the importance of direct and indirect interactions among 370

symbionts during symbiont community assembly (Graham 2008; Skelton, Creed & Brown 2015; 371

Skelton et al. 2016; Thomas et al. 2016). Crayfish growth and survivorship can be increased by 372


18

hosting branchiobdellidan worms (reviewed in Skelton et al. 2013). Several lines of evidence 373

suggest that the positive effect of worms on their host is mediated by interactions with microbial 374

communities that develop on the gills. Branchiobdellidans such as Cambarincola ingens often 375

enter the gill chamber and consume organic matter that accumulates therein, in addition to 376

grazing on the exoskeleton biofilm (Brown, Creed & Dobson 2002; Brown et al. 2012). 377

Accumulations of organic matter, particularly bacterial flocs, present a major challenge to 378

crayfish in their natural benthic environment (Bauer 1998; Rosewarne et al. 2014). Thus 379

researchers have hypothesized that some branchiobdellidans increase the growth and survival of 380

their hosts by consuming harmful organic matter, including bacteria flocs, from the gill epithelia 381

(Brown, Creed & Dobson 2002; Brown et al. 2012). Strong support for this hypothesis has come 382

from multiple studies demonstrating context-dependent outcomes in the interactions between 383

crayfish and their worms (Lee, Kim & Choe 2009; Brown et al. 2012; Skelton, Creed & Brown 384

2014; Thomas et al. 2016) 385

In this study, branchiobdellidans had a significant effect on the composition of carapace 386

microbial communities, but no detectable effect on microbial composition of the gills. Bacterial 387

richness was also unaffected on both gills and carapaces. The methods used in this study could 388

only detect changes in bacterial composition and not bacterial biomass, so it remains possible 389

that branchiobdellidans reduce bacterial biomass on the gills without specific effects on the 390

composition of bacterial communities. Because the gills were found in this study to be 391

dominated by a few taxa and perhaps influenced directly or indirectly by the host, any effects of 392

worms on gill bacterial communities is more likely to be quantitative than qualitative. In contrast 393

to the gills, microbial communities of the carapace appear to be more variable, and composed of 394

abundant opportunistic colonizers whose relative abundances are influenced by local factors such 395


19

as environment and branchiobdellidan presence. The effect of branchiobdellidans could be either 396

from direct grazing, indirect effects of grazing via resource alteration, or perhaps an effect of 397

adhesive compounds released by the worms’ duogland adhesive organs used to attach to their 398

hosts. Whatever the cause, the disparity between worm effects on gills and carapaces suggests 399

that gill microbial communities are less susceptible to alteration from local biological factors, 400

and that the symbiosis between crayfish and the bacterial communities of their gills is more 401

tightly coupled to host biology than environmental context. 402

Conclusions 403

Our goal for this study was to examine the effects of ecological processes on patterns of 404

microbial symbiont communities at multiple scales, from across watersheds to specific host 405

tissues. We provide the first characterization of the crayfish microbiome and identify taxa with 406

potential functional significance to host biology. We found significant variation in microbial 407

symbiont communities at each scale, i.e. among watersheds, among individual crayfish, and 408

among host tissues. We also provide experimental evidence that coinfection of a metazoan 409

symbiont may influence microbiome composition of some tissues. However, the vast majority of 410

variation in microbial composition and diversity was explained by host microsite, i.e., gills 411

versus carapaces. This result was consistent across DNA fingerprint analyses in our field survey 412

(TRFLP), and the results of sequencing from experimental samples. This result is also congruent 413

with microbiome work from other animals including humans (e.g. Grice & Segre 2011) and 414

suggests that the predominant ecological processes that dictate membership of the crayfish 415

microbiome operate at the finest spatial scales, where microorganisms meet tissue. Furthermore, 416

the microbial communities of the gills were found to be less responsive to local environmental 417

context and dominated by two bacterial families identified as symbionts of crayfish in this study 418


20

and symbionts of other aquatic animals elsewhere, suggesting that these organisms may 419

represent important specialized aquatic animal symbionts. Future work may uncover functional 420

significances of these microbial taxa that are closely associated with the gills of crayfish and the 421

skins of frogs and salamanders. 422

Acknowledgements: We sincerely thank Michael J. Thomas and Mathew Hedin for their help in 423

the field and laboratory. Allan Dickerman and the staff of the Virginia Tech Bioinformatics 424

Institute were critical partners for next-generation sequencing and analysis. This manuscript 425

benefited greatly from thoughtful comments from Michelle A. Jusino. Funding was provided by 426

the National Science Foundation (DEB-0949780 to BLB and DEB-0949823 to RPC), Virginia 427

Tech Organismal Biology and Ecology Interdisciplinary Grants (to JS and KMG), and the 428

Virginia Bioinformatics Institute and Fralin Life Science Institute Small Grants Program (to JS, 429

KMG, JTL and BLB). 430

Data accessibility: Data available from the Dryad Digital Repository: XXXXXX 431

432


21

Literature cited 433

Anderson, M.J. (2001) A new method for non‐parametric multivariate analysis of variance. 434 Austral Ecology, 26, 32-46. 435

Bäckhed, F., Ley, R.E., Sonnenburg, J.L., Peterson, D.A. & Gordon, J.I. (2005) Host-Bacterial 436 Mutualism in the Human Intestine. Science, 307, 1915-1920. 437

Bauer, R.T. (1998) Gill-cleaning mechanisms of the crayfish Procambarus clarkii (Astacidea: 438 Cambaridae): Experimental testing of setobranch function. Invertebrate Biology, 117, 439 129-143. 440

Bright, M. & Bulgheresi, S. (2010) A complex journey: transmission of microbial symbionts. 441 Nature Reviews Microbiology, 8, 218-230. 442

Brown, B., Creed, R.P. & Dobson, W.E. (2002) Branchiobdellid annelids and their crayfish 443

hosts: are they engaged in a cleaning symbiosis? Oecologia, 132, 250-255. 444

Brown, B.L., Creed, R.P., Skelton, J., Rollins, M.A. & Farrell, K.J. (2012) The fine line between 445 mutualism and parasitism: complex effects in a cleaning symbiosis demonstrated by 446

multiple field experiments. Oecologia, 170, 199-207. 447 Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., 448

Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Knights, 449 D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., 450 Reeder, J., Sevinsky, J.R., Tumbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., 451

Zaneveld, J. & Knight, R. (2010) QIIME allows analysis of high-throughput community 452 sequencing data. Nature Methods, 7, 335-336. 453

Clay, K. (2014) Defensive symbiosis: a microbial perspective. Functional Ecology, 28, 293-298. 454 Costello, E.K., Lauber, C.L., Hamady, M., Fierer, N., Gordon, J.I. & Knight, R. (2009) Bacterial 455

community variation in human body habitats across space and time. science, 326, 1694-456 1697. 457

Creed, R.P. (1994) Direct and indirect effects of crayfish grazing in a stream community. 458 Ecology, 2091-2103. 459

Creed, R.P. & Reed, J.M. (2004) Ecosystem engineering by crayfish in a headwater stream 460

community. Journal of the North American Benthological Society, 23, 224-236. 461 Dethlefsen, L., Eckburg, P.B., Bik, E.M. & Relman, D.A. (2006) Assembly of the human 462

intestinal microbiota. Trends in Ecology & Evolution, 21, 517-523. 463 Dethlefsen, L., McFall-Ngai, M. & Relman, D.A. (2007) An ecological and evolutionary 464

perspective on human–microbe mutualism and disease. Nature, 449, 811-818. 465 Dominguez-Bello, M.G., Costello, E.K., Contreras, M., Magris, M., Hidalgo, G., Fierer, N. & 466

Knight, R. (2010) Delivery mode shapes the acquisition and structure of the initial 467

microbiota across multiple body habitats in newborns. Proceedings of the National 468

Academy of Sciences, 107, 11971-11975. 469 Edgar, R.C. (2010) Search and clustering orders of magnitude faster than BLAST. 470

Bioinformatics, 26, 2460-2461. 471 Edgar, R.C. (2013) UPARSE: highly accurate OTU sequences from microbial amplicon reads. 472

Nature Methods, 10, 996. 473

Farrell, K.J., Creed, R.P. & Brown, B.L. (2014) Preventing overexploitation in a mutualism: 474 partner regulation in the crayfish–branchiobdellid symbiosis. Oecologia, 174, 501-510. 475


22

Fierer, N., Morse, J.L., Berthrong, S.T., Bernhardt, E.S. & Jackson, R.B. (2007) Environmental 476

controls on the landscape-scale biogeography of stream bacterial communities. Ecology, 477 88, 2162-2173. 478

Freese, H.M. & Schink, B. (2011) Composition and stability of the microbial community inside 479

the digestive tract of the aquatic crustacean Daphnia magna. Microbial Ecology, 62, 882-480 894. 481

Geyer, K.M., Altrichter, A.E., Van Horn, D.J., Takacs-Vesbach, C.D., Gooseff, M.N. & Barrett, 482 J.E. (2013) Environmental controls over bacterial communities in polar desert soils. 483 Ecosphere, 4, art127. 484

Gilbert, S.F., Sapp, J. & Tauber, A.I. (2012) A Symbiotic View of Life: We Have Never Been 485 Individuals. The Quarterly Review of Biology, 87, 325-341. 486

Graham, A.L. (2008) Ecological rules governing helminth–microparasite coinfection. 487 Proceedings of the National Academy of Sciences, 105, 566-570. 488

Grice, E.A. & Segre, J.A. (2011) The skin microbiome. Nature Reviews Microbiology, 9, 244-489 253. 490

Harris, R.N., Brucker, R.M., Walke, J.B., Becker, M.H., Schwantes, C.R., Flaherty, D.C., Lam, 491 B.A., Woodhams, D.C., Briggs, C.J., Vredenburg, V.T. & Minbiole, K.P.C. (2009) Skin 492

microbes on frogs prevent morbidity and mortality caused by a lethal skin fungus. ISME 493 J, 3, 818-824. 494

Helms, B., Loughman, Z.J., Brown, B.L. & Stoeckel, J. (2013) Recent advances in crayfish 495

biology, ecology, and conservation. Freshwater Science, 32, 1273-1275. 496 Hobbs, H.H.J., Holt, P.C. & Walton, M. (1967) The crayfishes and their epizootic ostracod and 497

branchiobdellid associates of the Mountain Lake, Virginia, Region. Poceedings of the 498 United States National Museum, 123, 1-84. 499

Holdich, D.M. & Crandall, K. (2002) Biology of freshwater crayfish. Blackwell Science Oxford. 500

Huttenhower, C., Gevers, D., Knight, R., Abubucker, S., Badger, J.H., Chinwalla, A.T., Creasy, 501

H.H., Earl, A.M., FitzGerald, M.G., Fulton, R.S., Giglio, M.G., Hallsworth-Pepin, K., 502 Lobos, E.A., Madupu, R., Magrini, V., Martin, J.C., Mitreva, M., Muzny, D.M., 503 Sodergren, E.J., Versalovic, J., Wollam, A.M., Worley, K.C., Wortman, J.R., Young, 504

S.K., Zeng, Q.D., Aagaard, K.M., Abolude, O.O., Allen-Vercoe, E., Alm, E.J., Alvarado, 505 L., Andersen, G.L., Anderson, S., Appelbaum, E., Arachchi, H.M., Armitage, G., Arze, 506

C.A., Ayvaz, T., Baker, C.C., Begg, L., Belachew, T., Bhonagiri, V., Bihan, M., Blaser, 507 M.J., Bloom, T., Bonazzi, V., Brooks, J.P., Buck, G.A., Buhay, C.J., Busam, D.A., 508

Campbell, J.L., Canon, S.R., Cantarel, B.L., Chain, P.S.G., Chen, I.M.A., Chen, L., 509 Chhibba, S., Chu, K., Ciulla, D.M., Clemente, J.C., Clifton, S.W., Conlan, S., Crabtree, 510 J., Cutting, M.A., Davidovics, N.J., Davis, C.C., DeSantis, T.Z., Deal, C., Delehaunty, 511 K.D., Dewhirst, F.E., Deych, E., Ding, Y., Dooling, D.J., Dugan, S.P., Dunne, W.M., 512

Durkin, A.S., Edgar, R.C., Erlich, R.L., Farmer, C.N., Farrell, R.M., Faust, K., 513 Feldgarden, M., Felix, V.M., Fisher, S., Fodor, A.A., Forney, L.J., Foster, L., Di 514 Francesco, V., Friedman, J., Friedrich, D.C., Fronick, C.C., Fulton, L.L., Gao, H.Y., 515

Garcia, N., Giannoukos, G., Giblin, C., Giovanni, M.Y., Goldberg, J.M., Goll, J., 516 Gonzalez, A., Griggs, A., Gujja, S., Haake, S.K., Haas, B.J., Hamilton, H.A., Harris, 517 E.L., Hepburn, T.A., Herter, B., Hoffmann, D.E., Holder, M.E., Howarth, C., Huang, 518 K.H., Huse, S.M., Izard, J., Jansson, J.K., Jiang, H.Y., Jordan, C., Joshi, V., Katancik, 519 J.A., Keitel, W.A., Kelley, S.T., Kells, C., King, N.B., Knights, D., Kong, H.D.H., Koren, 520 O., Koren, S., Kota, K.C., Kovar, C.L., Kyrpides, N.C., La Rosa, P.S., Lee, S.L., Lemon, 521


23

K.P., Lennon, N., Lewis, C.M., Lewis, L., Ley, R.E., Li, K., Liolios, K., Liu, B., Liu, Y., 522

Lo, C.C., Lozupone, C.A., Lunsford, R.D., Madden, T., Mahurkar, A.A., Mannon, P.J., 523 Mardis, E.R., Markowitz, V.M., Mavromatis, K., McCorrison, J.M., McDonald, D., 524 McEwen, J., McGuire, A.L., McInnes, P., Mehta, T., Mihindukulasuriya, K.A., Miller, 525

J.R., Minx, P.J., Newsham, I., Nusbaum, C., O'Laughlin, M., Orvis, J., Pagani, I., 526 Palaniappan, K., Patel, S.M., Pearson, M., Peterson, J., Podar, M., Pohl, C., Pollard, K.S., 527 Pop, M., Priest, M.E., Proctor, L.M., Qin, X., Raes, J., Ravel, J., Reid, J.G., Rho, M., 528 Rhodes, R., Riehle, K.P., Rivera, M.C., Rodriguez-Mueller, B., Rogers, Y.H., Ross, 529 M.C., Russ, C., Sanka, R.K., Sankar, P., Sathirapongsasuti, J.F., Schloss, J.A., Schloss, 530

P.D., Schmidt, T.M., Scholz, M., Schriml, L., Schubert, A.M., Segata, N., Segre, J.A., 531 Shannon, W.D., Sharp, R.R., Sharpton, T.J., Shenoy, N., Sheth, N.U., Simone, G.A., 532 Singh, I., Smillie, C.S., Sobel, J.D., Sommer, D.D., Spicer, P., Sutton, G.G., Sykes, S.M., 533 Tabbaa, D.G., Thiagarajan, M., Tomlinson, C.M., Torralba, M., Treangen, T.J., Truty, 534

R.M., Vishnivetskaya, T.A., Walker, J., Wang, L., Wang, Z.Y., Ward, D.V., Warren, W., 535 Watson, M.A., Wellington, C., Wetterstrand, K.A., White, J.R., Wilczek-Boney, K., Wu, 536

Y.Q., Wylie, K.M., Wylie, T., Yandava, C., Ye, L., Ye, Y.Z., Yooseph, S., Youmans, 537 B.P., Zhang, L., Zhou, Y.J., Zhu, Y.M., Zoloth, L., Zucker, J.D., Birren, B.W., Gibbs, 538

R.A., Highlander, S.K., Methe, B.A., Nelson, K.E., Petrosino, J.F., Weinstock, G.M., 539 Wilson, R.K., White, O. & Human Microbiome Project, C. (2012) Structure, function and 540 diversity of the healthy human microbiome. Nature, 486, 207-214. 541

Jost, L. (2006) Entropy and diversity. Oikos, 113, 363-375. 542 Kuczynski, J., Costello, E.K., Nemergut, D.R., Zaneveld, J., Lauber, C.L., Knights, D., Koren, 543

O., Fierer, N., Kelley, S.T. & Ley, R.E. (2010) Direct sequencing of the human 544 microbiome readily reveals community differences. Genome Biol, 11, 210. 545

Kueneman, J.G., Parfrey, L.W., Woodhams, D.C., Archer, H.M., Knight, R. & McKenzie, V.J. 546

(2014) The amphibian skin-associated microbiome across species, space and life history 547

stages. Molecular ecology, 23, 1238-1250. 548 Lauber, C.L., Hamady, M., Knight, R. & Fierer, N. (2009) Pyrosequencing-based assessment of 549

soil pH as a predictor of soil bacterial community structure at the continental scale. 550

Applied and Environmental Microbiology, 75, 5111-5120. 551 Lee, J.H., Kim, T.W. & Choe, J.C. (2009) Commensalism or mutualism: conditional outcomes in 552

a Branchiobdellidae crayfish symbiosis. Oecologia, 159, 217-224. 553 Loudon, A.H., Woodhams, D.C., Parfrey, L.W., Archer, H., Knight, R., McKenzie, V. & Harris, 554

R.N. (2014) Microbial community dynamics and effect of environmental microbial 555 reservoirs on red-backed salamanders (Plethodon cinereus). Isme Journal, 8, 830-840. 556

Masella, A.P., Bartram, A.K., Truszkowski, J.M., Brown, D.G. & Neufeld, J.D. (2012) 557 PANDAseq: PAired-eND Assembler for Illumina sequences. Bmc Bioinformatics, 13, 7. 558

McKenzie, V.J., Bowers, R.M., Fierer, N., Knight, R. & Lauber, C.L. (2012) Co-habiting 559 amphibian species harbor unique skin bacterial communities in wild populations. Isme 560 Journal, 6, 588-596. 561

Mihaljevic, J.R. (2012) Linking metacommunity theory and symbiont evolutionary ecology. 562 Trends in Ecology & Evolution, 27, 323-329. 563

Mitchell, K.R. & Takacs-Vesbach, C.D. (2008) A comparison of methods for total community 564 DNA preservation and extraction from various thermal environments. Journal of 565 Industrial Microbiology & Biotechnology, 35, 1139-1147. 566


24

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, 567

G.L., Solymos, P., Henry, M., Stevens, H. & Wagner, H. (2013) vegan: Community 568 Ecology Package. 569

Owen, C.L., Bracken-Grissom, H., Stern, D. & Crandall, K.A. (2015) A synthetic phylogeny of 570

freshwater crayfish: insights for conservation. Philosophical Transactions of the Royal 571 Society of London B: Biological Sciences, 370, 20140009. 572

Pedersen, A.B. & Fenton, A. (2007) Emphasizing the ecology in parasite community ecology. 573 Trends in Ecology & Evolution, 22, 133-139. 574

Peerakietkhajorn, S., Tsukada, K., Kato, Y., Matsuura, T. & Watanabe, H. (2015) Symbiotic 575

bacteria contribute to increasing the population size of a freshwater crustacean, Daphnia 576 magna. Environmental microbiology reports, 7, 364-372. 577

Pequeux, A. (1995) Osmotic regulation in crustaceans. Journal of Crustacean Biology, 15, 1-60. 578 Qi, W., Nong, G., Preston, J.F., Ben-Ami, F. & Ebert, D. (2009) Comparative metagenomics of 579

Daphnia symbionts. BMC genomics, 10, 1. 580 Rosenberg, E. (2014) The Family Chitinophagaceae. The Prokaryotes: Other Major Lineages of 581

Bacteria and The Archaea (eds E. Rosenberg, E.F. DeLong, S. Lory, E. Stackebrandt & 582 F. Thompson), pp. 493-495. Springer Berlin Heidelberg, Berlin, Heidelberg. 583

Rosewarne, P., Svendsen, J., Mortimer, R.G. & Dunn, A. (2014) Muddied waters: suspended 584 sediment impacts on gill structure and aerobic scope in an endangered native and an 585 invasive freshwater crayfish. Hydrobiologia, 722, 61-74. 586

Rousk, J., Baath, E., Brookes, P.C., Lauber, C.L., Lozupone, C., Caporaso, J.G., Knight, R. & 587 Fierer, N. (2010) Soil bacterial and fungal communities across a pH gradient in an arable 588

soil. ISME J, 4, 1340-1351. 589 Scheuring, I. & Yu, D.W. (2012) How to assemble a beneficial microbiome in three easy steps. 590

Ecology letters, 15, 1300-1307. 591

Shafquat, A., Joice, R., Simmons, S.L. & Huttenhower, C. (2014) Functional and phylogenetic 592

assembly of microbial communities in the human microbiome. Trends in Microbiology, 593 22, 261-266. 594

Skelton, J., Creed, R.P. & Brown, B.L. (2014) Ontogenetic shift in host tolerance controls 595

initiation of a cleaning symbiosis. Oikos, 123, 677-686. 596 Skelton, J., Creed, R.P. & Brown, B.L. (2015) A symbiont’s dispersal strategy: Condition-597

dependent dispersal underlies predictable variation in direct transmission among hosts. 598 Proceedings of the Royal Society B: Biological Sciences, 282(1819). 599

Skelton, J., Doak, S., Leonard, M., Creed, R.P. & Brown, B.L. (2016) The rules for symbiont 600

community assembly change along a mutualism‐parasitism continuum. Journal of Animal 601 Ecology, 85, 843-853. 602

Skelton, J., Farrell, K.J., Creed, R.P., Williams, B.W., Ames, C., Helms, B.S., Stoekel, J. & 603

Brown, B.L. (2013) Servants, scoundrels, and hitchhikers: current understanding of the 604 complex interactions between crayfish and their ectosymbiotic worms 605 (Branchiobdellida). Freshwater Science, 32, 1345-1357. 606

Statzner, B., Fievet, E., Champagne, J.Y., Morel, R. & Herouin, E. (2000) Crayfish as 607 geomorphic agents and ecosystem engineers: biological behavior affects sand and gravel 608 erosion in experimental streams. Limnology and Oceanography, 1030-1040. 609

Statzner, B., Peltret, O. & Tomanova, S. (2003) Crayfish as geomorphic agents and ecosystem 610

engineers: effect of a biomass gradient on baseflow and flood‐induced transport of gravel 611 and sand in experimental streams. Freshwater biology, 48, 147-163. 612


25

Taylor, C.A., Schuster, G.A., Cooper, J.E., DiStefano, R.J., Eversole, A.G., Hamr, P., Hobbs III, 613

H.H., Robison, H.W., Skelton, C.E. & Thoma, R.F. (2007) A reassessment of the 614 conservation status of crayfishes of the United States and Canada after 10+ years of 615 increased awareness. Fisheries, 32, 372-389. 616

Thomas, M.J., Creed, R.P., Skelton, J. & Brown, B.L. (2016) Ontogenetic shifts in a freshwater 617 cleaning symbiosis: consequences for hosts and their symbionts. Ecology. 618

Usio, N. & Townsend, C.R. (2002) Functional significance of crayfish in stream food webs: 619 roles of omnivory, substrate heterogeneity and sex. Oikos, 98, 512-522. 620

Usio, N. & Townsend, C.R. (2004) Roles of crayfish: consequences of predation and 621

bioturbation for stream invertebrates. Ecology, 85, 807-822. 622 Walke, J.B., Becker, M.H., Loftus, S.C., House, L.L., Cormier, G., Jensen, R.V. & Belden, L.K. 623

(2014) Amphibian skin may select for rare environmental microbes. ISME J. 624 Walke, J.B., Becker, M.H., Loftus, S.C., House, L.L., Teotonio, T.L., Minbiole, K.P. & Belden, 625

L.K. (2015) Community Structure and Function of Amphibian Skin Microbes: An 626 Experiment with Bullfrogs Exposed to a Chytrid Fungus. PLoS One, 10, e0139848. 627

628

629


26

Figure 1: Conceptual diagram showing multi-level filtering of microbial symbiont community 630

during symbiont community assembly. Colonizing potential symbionts are filtered by habitat 631

characteristics such as water chemistry and pH at the host habitat level (a), by host characteristics 632

at the host-habitat interface (b), and by symbiont interactions and tissue-specific characteristics 633

at the host microsite level (c). The community structure observed at any microsite within the host 634

is the product of a series of ecological filters that operate at scales from geographic to cellular. 635

636

637


27

Figure 2. Field survey results depicting the relationship between microbial diversity originating 638

from crayfish (gills[grey symbols]; carapace[black symbols]) and surrounding environment 639

(substrate) from four sampling sites, ± 1 SE. Diversity is shown as the effective number of 640

equally abundant unique fragment lengths, derived from Simpson’s Index of relativized TRFLP 641

data. Grey reference shows 1:1 relationship. Diversity of carapace was typically higher than, and 642

positively correlated with substrate diversity. Gill diversity was typically lower than, and 643

uncorrelated with substrate diversity. 644

645


28

Figure 3: Non-metric multidimensional scaling (NMDS) of field survey results showing 646

relationships among the bacterial biofilm community of gills (squares), carapaces (circles) and 647

substrate (shaded ellipses) from four locations; red = “Sinking Creek”, blue = “Big Stoney”, 648

green = “Spruce run”, grey = “Toms Creek”. Symbols represent centroid (mean) of each group. 649

Ellipses represent 95% confidence envelope for centroid of substrate samples. Community 650

composition of the carapaces was less variable, but within the range of local substrate. Gill 651

composition was variable at some sites, but always distinct from substrate. 652

653


29

Figure 4. Comparison of microbial communities of the crayfish carapace and gills. (left) Rank 654

abundance curves for the 100 most frequently observed taxa showed that carapace communities 655

were more even than those of the gills, and gills were largely dominated by few taxa. (right) Heat 656

map of showing the log counts of reads for 30 most frequently observed OTUs revealed obvious 657

compositional differences between gills and carapaces. Columns are arranged by similarity using 658

hierarchical clustering. Many taxa were equally abundant on the carapace, however, the gills 659

were dominated by two taxa, Comamonadacea and Chitnophagacea. Letters before taxa indicate 660

finest taxonomic rank resolved from bioinformatics pipeline (e.g., g=genus, f=family, o=order). 661

662


30

Figure 5. Non-metric multidimensional scaling ordination showing significant effects of 663

branchiobdellidan treatment on carapace (right; stress = 0.14) but not gill (left; stress = 0.08) 664

microbial communities. 665

666

667


31

Table 1: Linear regression for effects of average substrate diversity and crayfish tissue type on 668

bacterial diversity. The overall model was highly significant (F3, 37 = 30.21, p < 0.001). 669

Coefficients Estimate t statistic p value

Intercept 10.25 3.685 < 0.001*

Sample type (Gills) -2.91 -0.718 0.477

Mean substrate diversity 0.521 2.507 0.017*

Sample type × substrate diversity -0.521 0.300 0.090

670

671


32

Table 2: Linear regression for effects of branchiobdellidan worm treatment and crayfish tissue 672

type on bacterial diversity. The overall model was highly significant (F3, 28 = 80.13, p < 0.001). 673

Coefficients Estimate t statistic p value

Intercept 36.18 18.045 < 0.001*

Worm treatment 0.187 0.254 0.466

Sample type (Gills) -30.179 -9.663 < 0.001*

Sample type × worm treatment --0.258 -0.658 0.516

674

675


Multi-scale ecological filters shape the crayfish … 1 Major article: Multi-scale ecological...

Documents

Transcript of Multi-scale ecological filters shape the crayfish … 1 Major article: Multi-scale ecological...